Novel method for online wear estimation of centrifugal pumps using multi-fidelity modeling

Integrated online condition monitoring services are becoming more and more a key for industrial system operators to monitor and extend the lifetime of their machinery. Therefore, system developers need to efficiently provide system operators so-called digital twins which can be executed during operation and provide the system operator an impression of the health state of the system. Pump curves of centrifugal pumps are a simple but effective model the pump and typically determined in the design phase of the pump. During operation the pump curves change due to different types of wear which can occur. This article addresses a method how the pump health can be estimated during operation without the necessity of a system stop.